The present application is a revision of an earlier application requesting four years of support to conduct a longitudinal follow-up study of a recently completed random sample survey of 2052 black and white adolescents, aged 13 to 19. An 81% response rate was obtained. A conceptual model was presented to guide this research in which drinking in heterosexual social or sexual situations was hypothesized to increase the likelihood of engaging in high risk sexual behaviors. It was further proposed that drinking in sexual or potentially sexual situations may occur for one of two reasons: to reduce threat or manage discomfort in situations experienced as threatening (relief drinking), or to enhance positive affect in situations experienced as non-threatening (enhancement drinking). In turn, drinking for relief vs. enhancement motives was hypothesized to reflect different underlying etiologic processes. Analyses of our cross-sectional data provided strong support for key tenets of the proposed model, indicating that alcohol use in social/sexual situations may be a unique risk factor for high risk sexual behavior, and that drinking in these situations can be usefully understood within the proposed cognitive motivational framework. At the same time, however, we also found evidence that at least part of the association between drinking and risky sexual behaviors can be accounted for by more general lifestyle and personality variables. We therefore propose to continue analyses and publication of our cross- sectional data, and to conduct a four-year longitudinal follow-up of these adolescents. Broadly defined, the objectives for the proposed follow-up include: (l) implementation of both methodological and conceptual refinements to the event-level approach used to examine the alcohol-risky sexual behavior link in our cross-sectional survey, and (2)expansion of our focus to examine non-causal as well as reciprocal and reverse causal processes that may underlie the association between drinking and risky sexual behavior. In addition, gender, race, and age differences in these relationships will be systematically explored. By embracing a range of plausible causal and non-causal models, we hope to contribute to a fuller explication of the multiple processes likely to underlie and account for the relationship between alcohol use and sexual risk behaviors.